Forum Replies Created

Viewing 15 posts - 16 through 30 (of 166 total)
  • Author
    Posts
  • in reply to: Debugging the tricopter mini racer #27793
    rue
    Participant

    Need a little tuning hint – “normal” flight behavior is quite nice, but when I do a full throttle climb, the yaw servo starts to oscillate. Would I have to lower P or rather increase D?

    Running on Betaflight 2.6.1 / LUX / 1kHz / 4S / David’s electronics

    Thanks!!

    in reply to: vibrations w/o props normal? #27728
    rue
    Participant

    Thanks, David. The copter was sitting on a table – will try to suspend it and re-run the test.

    in reply to: Idea for TriFlight (aka Naze32) Frame V2? #27658
    rue
    Participant

    The transponder is ranked #2 on my personal list of things I’ll surely NOT NEED thanks to my habit of rather hitting than avoiding lokal trees and my “seriously non-pro” flying skills (#1 btw is all this “seriously pro / yaw dropping / designed for racing” lingo on the web page).

    To me, the board alignment looks o.k. on the photo as the MPU9250 is mounted such that the USB port looks sideways with default orientation in CF.

    Nice for building quads would have been to route the motor outs to the corners of the board.

    in reply to: Idea for TriFlight (aka Naze32) Frame V2? #27645
    rue
    Participant

    To bridge some waiting time πŸ˜‰ Dominik Clifton has just released a new variant of his F3-based boards, the SPRacing F3 Evo, which imho gets quite a bunch of things right (incl. the price tag): SPI connected Gyro/Acc/Mag, CFCard slot, regular through-holes for the ports. Ticks all check-boxes on my list…

    in reply to: Debugging the tricopter mini racer #27603
    rue
    Participant

    Not sure why @lauka has changed the default. Advantages of higher looptimes are the reduced sensitivity to vibration induced artifacts due to higher sampling rate and better options to tweak filters, with the “built in” Naze of the RCE frame where there’s no option to stick some foam underneath the board this actually might come quite handy. Have been following betaflight more closely…

    in reply to: Debugging the tricopter mini racer #27599
    rue
    Participant

    > It’s a pity when outdated hardware is a performance limiting factor.

    You’re right – but my point was that if you’re striving for a 1k looptime on LUX the F1 is quite happy to deliver. Going lower than that requires you to deactivate ACC on F1 (not necessary on F3). Check out the BF wiki and thread about the achievable looptimes on different targets.

    For new controllers it’s absolutely clear that F3 is the way to go.

    in reply to: Debugging the tricopter mini racer #27571
    rue
    Participant

    I second @lauka – luxfloat is working fine on f1 boards and actually quite some cleanup has been performed in betaflight to remove some unnecessary floating point math. So unless you’re unhappy with 1 kHz PID loop your f1 boards will provide for happy flying for quite a while πŸ˜‰

    BTW – one way to skip some floating point calculation is to avoid using align_board_pitch and use the sensor alignment instead:

    set align_gyro = CW0FLIP
    set align_acc = CW0FLIP

    None the less, I’m also very much looking forward to an F3 based frame πŸ™‚

    in reply to: throttle issues #27118
    rue
    Participant

    If you find throttle too responsive you could give it some expo via the Receiver Tab in the CleanFlight Configurator…

    rue
    Participant

    I’ve also had issues in the stabilized modes and found some cleanflight fixes that had not been merged to the betaflight/triflight base (as of January, that is). See here

    Not sure if these fixes have been merged on the way to the current version of triflight. Let’s see if @lauka has a hint on this.

    in reply to: did I kill my naze32? #26601
    rue
    Participant

    Hi Allenhat – did you try the sequence of steps indicated in the “recovery / lost communication” box on Cleanflight’s flash tab? It’s basically what David suggests in the second part of his post, but every step is important. Also really solder-bridge the boot pads, trying to short it with pincers/screwdrivers etc. usually does not work.

    Best of luck!

    in reply to: First Build – Arms Don't Fit Through Screws in Motor #26252
    rue
    Participant

    Yep – it’s surprising how much mud you can capture with just one boom πŸ™‚

    in reply to: CLI Error, Corrupted Board? #26245
    rue
    Participant

    @deepnine – correct πŸ˜‰ Anywhere close?

    in reply to: CLI Error, Corrupted Board? #26244
    rue
    Participant

    Deepnine & habahukk are right – you’d be missing out on serious fun if you shelf the poor tri now. David’s setup is actually a very reliably one with all components (especially motors/escs/battery) specced nicely for operating at the upper end within (!) their specs. So smoking ESCs and smelly motors stay the exception.

    Only one component in the overall system failed shockingly often (causing punctured lipos, bent / ripped wires, abused servo gears, a motor, a CF boom, even the antennas for my frsky rx) – the moron behind the sticks seems to have a passion for not-avoiding trees, sometimes tries out new things he’s not capable of coordinating or, for no apparent reason whatsoever, loses orientating flying LOS and slams the poor tri into sth hard and nasty. Need to get a better version…

    in reply to: BEC pad no longer attached to frame #26243
    rue
    Participant

    Pls send picture of what you have done – I’m powering the whole thing through an external BEC connected to an arbitrary open slot on the pin header, just as described by the others above.

    Just to make sure: did you have the battery connected? Only the controller board itself is powered through the USB connection, it does not feed through to the attached components –> you need the battery as soon as you need to power anything on top of the naze.

    in reply to: CLI Error, Corrupted Board? #26231
    rue
    Participant

    The two adjacent pads BOOT and BOOT/3.3V

Viewing 15 posts - 16 through 30 (of 166 total)